We have recently observed that rats exposed to alcohol for three consecutive days via indwelling intragastric cannulae (Model A) showed the expected initial rise in plasma ACTH and corticosterone levels, but that upon a re-challenge with the drug 3-12 days later, these hormone responses were significantly blunted. Under Specific Aim 1, we will test hypothesis related to the mechanisms responsible for the decreased HPA axis in this model of short-term alcohol treatment. We will first determine whether pituitary responsiveness to CRF and/or vasopressin is reduced. If this is the case, we will test the hypothesis that steady-state MRNA levels of pituitary CRF-R1 and/or POMC are decreased, and/or that the pituitary is more sensitive to glucocorticoid feedback. Using transcript levels of the immediate early gene NGFI-B, we will also determine whether hypothalamic neuronal activation is blunted during a second alcohol challenge. If they are reduced, we will test the hypothesis that this is accompanied by decreased CRF synthesis in the hypothalamus, possibly via increased sensitivity to CRF neurons to glucocorticoid feedback. In parallel studies, we will also use a model of chronic alcohol treatment (Model B), in which rats are exposed to intermittent alcohol vapors for 2 weeks, then re-challenged with this drug following a period of protracted abstinence. We will first determine whether animals of Model B also show neuroendocrine tolerance when re-exposed to alcohol or whether, as observed with other long-term stress paradigms, they will exhibit an exacerbated response of their HPA axis. In either case, the mechanisms responsible for these changes will be investigated. Experiments carried out under Specific Aim 2, on the other hand, will use knock-out mice to investigate the importance of CRF and its receptors on the response of the HPA axis to alcohol. In CRF-deficient mice, we will test the hypothesis that CRF is the sole mediator of this response. In mice homozygous for a null CRF-R1 allele, we will test the hypothesis that pituitary and/or hypothalamic CRF-R1 participate in alcohol-induced ACTH release. Finally in CRF-R/2 knock-out mice, we will test the hypothesis that pituitary and/or hypothalamic CRF-R2 are not necessary for this ACTH response.